80 BULLETIN 1036, U. S. DEPARTMENT OF AGRICULTURE. 
thing is to obtain the cheapest material that will meet the conditions. 
Benzene and chloroform would answer the requirements so far as 
physical and chemical conditions are concerned, and on drying they 
would evaporate just as water does; but their use is, of course, out 
of the question on account of their great cost. Wood-preservers 
are practically confined by cost to such crude raw oils as crude 
petroleum, coal-tar, or water-gas tar. Of these three, coal-tar and 
water-gas tar are better solvents of creosote than is crude petroleum. 
Of the two tars, water-gas tar seems for the following reasons to fulfill 
more nearly all the requirements of a diluent: 
(1) It contains an exceedingly small amount of suspended matter, 
whereas coal tar, as a minimum, has about 3 or 4 per cent. The 
viscosity of water-gas tar is usually less than that of coal tar. A 
mixture of water-gas tar and creosote may, therefore, be expected to 
penetrate more rapidly than an equal mixture of coal tar and coal-tar 
creosote. 
(2) It may be used in its raw state when it is free from water. 
Coal tar containing ammonia was found by Bolton to be injurious to 
wood; therefore it must be refined in order to remove the ammonia. 
(3) Crude water-gas tar, water free, is cheaper than refined coal 
tar. 
Inasmuch as service records {27) show that the failures so far 
.experienced in ties, poles, and other timbers have been the result 
of mechanical wear, checking, and similar causes, and not of the 
failure of the preservative, it is apparent that the limit of the life of 
coal-tar creosote has not been reached in this kind of service. In 
other words, the factor of safety is probably very much larger than it 
need be. It would seem justifiable in such conditions to dilute the 
creosote with some cheaper material, such as tar, to make the life 
of the preservative and the mechanical life of the wood more nearly 
equal. The best kind of oil to use for dilution under such conditions 
would be the low-boiling creosotes, because these contain the greatest 
amount of toxic materials. If the theory of the mechanism of pre- 
servative action suggested in this bulletin is correct, these low- 
boiling creosotes probably need more high-boiling materials to retard 
their too rapid solution. 
PROPERTIES OF TAR-CREOSOTE SOLUTIONS. 
But little can be said about the properties of tar-creosote solutions 
except that they should be intermediate between the same properties 
of creosote and tar. A small amount of work has, however, been 
done and the results of it are given below: 
SPECIFIC GRAVITY. 
The addition of a heavier substance, such as tar to creosote will, 
of course, increase the average specific gravity of the solution. This 
